Abstract: A method and apparatus for analyzing potential inclusion dependencies between columns of database tables is provided. A client issues a command, e.g., a SQL statement to a database server. The command is processed at the database server without the performance of a join. Result data that indicates any exact inclusion dependencies or approximate inclusion dependencies between columns of database tables identified in the command is transmitted from the database server to the client. In processing the command, values stored in columns of database tables are converting into integer representations that require fewer resources to analyze. Disclosed techniques provide for the use of an index on a single database table rather than defining an index on each database table, which can be impractical. Further, the performance of only one pass on the data is required to identify any inclusion dependencies between columns of database tables.

Abstract: A computer automatically reads each object in a metadata that is descriptive of a database. Then the computer automatically applies a number of rules, to check if the object triggers a rule. If so, then the computer automatically changes the metadata as per an action identified by the triggered rule. After all rules are applied to the object, and after all objects in the metadata are processed, the resulting metadata is used to change the database's structure. The objects may be read from the metadata in an order (e.g. hierarchy) specified in meta-metadata that is descriptive of the metadata.